Dyson, Inc. v. SharkNinja Operating LLC

• Court: District Court, N.D. Illinois
• Decided: March 31, 2019
• Docket: 1:14-cv-09442
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF ILLINOIS EASTERN DIVISION

DYSON, INC., ) ) Plaintiff, ) 14 C 9442 ) vs. ) Judge Gary Feinerman ) SHARKNINJA OPERATING LLC and SHARKNINJA ) SALES COMPANY, ) ) Defendants. ) MEMORANDUM OPINION AND ORDER Dyson, Inc. brought this suit in November 2014 against SharkNinja Operating LLC and SharkNinja Sales Company (together, “SharkNinja”), alleging false advertising in violation of the Lanham Act, 15 U.S.C. §§ 1051 et seq. Doc. 1. (At the time, the SharkNinja entities were known as Euro-Pro Operating LLC and Euro-Pro Sales Company. Doc. 156.) Dyson sought a preliminary injunction, Doc. 8, and SharkNinja moved to dismiss, transfer, or stay the case, Doc. 18. The court denied both motions. Docs. 63, 65 (Gottschall, J.) (reported at 2014 WL 7366030 (N.D. Ill. Dec. 22, 2014)); Docs. 117-118 (Gottschall, J.) (reported at 2015 WL 1120006 (N.D. Ill. Mar. 10, 2015)). After Dyson amended its complaint, Doc. 153, the parties cross-moved for summary judgment, Docs. 313, 322, 357. The court granted partial summary judgment to Dyson as to SharkNinja’s affirmative defenses and partial summary judgment to SharkNinja as to all of Dyson’s claims except its literal falsity claim for the period from August 2014 through December 2014. Doc. 424 (Gottschall, J.) (reported at 259 F. Supp. 3d 816 (N.D. Ill. 2017)). The case was then reassigned to the undersigned judge’s calendar. Doc. 449. After a two-week trial, the jury returned a general verdict finding SharkNinja liable and that its false advertising was willful, and awarding Dyson an accounting of SharkNinja’s profits in the amount of $16,410,681. Doc. 737. The court entered judgment consistent with the verdict. Doc. 738. During trial, SharkNinja moved under Civil Rule 50(a) for judgment as a matter of law on all of Dyson’s claims both at the close of Dyson’s case, Doc. 707; Doc. 804 at 99-100, and at

the close of the evidence, Doc. 717; Doc. 808 at 110. For its part, Dyson moved under Rule 50(a) for judgment as a matter of law on one of its liability theories and on one of SharkNinja’s requested deductions from Dyson’s damages. Doc. 716. The court denied those motions. Doc. 736. SharkNinja now renews its motion for judgment as a matter of law under Rule 50(b), Doc. 754, and moves in the alternative under Rule 59 for a new trial or a remittitur, Doc. 756. For its part, Dyson moves under 15 U.S.C. § 1117(a) for a finding that it is entitled to recover its attorney fees because this case is “exceptional,” Doc. 742, and under Rule 59(e) to amend the judgment to include an award of prejudgment interest, Doc. 749. Dyson’s prejudgment interest motion is granted, and the other motions are denied. Background

At trial, Dyson claimed that SharkNinja engaged in false advertising from August 25, 2014 through December 15, 2014 by asserting in advertisements that independent lab tests conducted under the ASTM F608 standard proved that SharkNinja’s NV650 vacuum cleaned carpets better than Dyson’s DC65 vacuum. Doc. 809 at 48. Dyson advanced two theories of falsity. First, Dyson contended that SharkNinja’s assertion that “independent lab tests proved” that its product cleaned carpets better than Dyson’s—which the court will call the “independence representation”—was false, either because the tests at issue, which were conducted for SharkNinja by nonparty testing firm Intertek, “were not conducted in an independent manner,” or because they “were not conducted in compliance with the ASTM F608 standard.” Id. at 49- 50. Second, Dyson contended that part of a SharkNinja infomercial was false in that it combined a voiceover about independent tests with a bar graph—which the parties call the “grams graphic”—displaying data from internal tests finding that on one type of carpet, SharkNinja’s NV650 picked up 42 grams of dirt to the Dyson DC65’s 37 grams. Doc. 796 at 106-107;

Doc. 809 at 79; Trial Exh. 12 at 6:07-6:18, 10:04-10:15. As noted, the jury returned a general verdict finding SharkNinja liable and that its false advertising was willful, and awarding Dyson an accounting of SharkNinja’s profits in the amount of $16,410,681. Doc. 737. Discussion I. SharkNinja’s Rule 50(b) Renewed Motion for Judgment as a Matter of Law Under Rule 50(b), judgment as a matter of law is proper if “a reasonable jury would not have a legally sufficient evidentiary basis” to support a verdict for the nonmovant. Fed. R. Civ. P. 50(a)(1), (b); see Martinez v. City of Chicago, 900 F.3d 838, 844 (7th Cir. 2018). The court’s task on a Rule 50(b) motion “is to decide whether a highly charitable assessment of the evidence supports the jury’s verdict or if, instead, the jury was irrational to reach its conclusion.” Thorne v. Member Select Ins. Co., 882 F.3d 642, 644 (7th Cir. 2018) (internal quotation marks

omitted). “[A] verdict supported by no evidence or a mere scintilla of evidence will not stand.” Martin v. Milwaukee Cnty., 904 F.3d 544, 550 (7th Cir. 2018). The court on a Rule 50(b) motion views the evidence at trial in the light most favorable to Dyson, the nonmovant, and draws all reasonable inferences in Dyson’s favor. See ibid. The court “may not make credibility determinations or weigh the evidence.” Martinez, 900 F.3d at 844 (internal quotation marks omitted). “[A]lthough the court should review the record as a whole, it must disregard all evidence favorable to the moving party that the jury is not required to believe.” Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 151 (2000); see also Robinson v. Perales, 894 F.3d 818, 833 (7th Cir. 2018) (same). “That is, the court should give credence to the evidence favoring the nonmovant as well as that evidence supporting the moving party that is uncontradicted and unimpeached, at least to the extent that that evidence comes from disinterested witnesses.” Reeves, 530 U.S. at 151 (internal quotation marks omitted). SharkNinja argues that it deserves judgment as a matter of law because Dyson did not

present evidence sufficient to support its claim that the independence representation was false— whether because the tests were not conducted independently, or because they were not conducted in compliance with ASTM F608—or its claim that the grams graphic was false. Doc. 754 at 7- 15. As to the grams graphic, SharkNinja argues further that Dyson failed to prove materiality or injury. Id. at 15-16. Finally, SharkNinja argues that even if it violated the Lanham Act, Dyson is not entitled to any monetary recovery because Dyson failed to prove that SharkNinja profited from its violations. Id. at 16-20. A. SharkNinja’s Failure to Preserve Its Challenge to the Sufficiency of the Evidence Regarding the Grams Graphic Dyson argues that SharkNinja’s challenges to the evidence regarding ASTM F608 compliance and the grams graphic are procedurally barred because it failed to preserve them in its Rule 50(a) motion. Doc. 768 at 7. SharkNinja disagrees, arguing that it preserved its challenge to the evidence regarding ASTM F608 compliance in its Rule 50(a) motion and its challenge to the evidence regarding the grams graphic in both its reply in support of that motion and its response to Dyson’s Rule 50(a) motion. Doc. 774 at 7-8. Because Rule 50(b) permits renewal after judgment only of a Rule 50(a) motion made

during trial, a Rule 50(b) movant “may raise only the grounds it advanced in the pre-verdict [Rule 50(a)] motion.” Andy Mohr Truck Ctr., Inc. v. Volvo Trucks N. Am., 869 F.3d 598, 604 (7th Cir. 2017); accord Avery v. City of Milwaukee, 847 F.3d 433, 438 (7th Cir. 2017) (similar). “Thus, if a party raises a new argument in its Rule 50(b) motion that was not presented in the Rule 50(a) motion, the non-moving party can properly object.” Andy Mohr Truck Ctr., 869 F.3d at 605 (internal quotation marks omitted). SharkNinja preserved its sufficiency of the evidence challenge as to the ASTM F608 compliance issue, but not as to the grams graphic. As to the ASTM F608 compliance issue, SharkNinja’s Rule 50(a) motion argued that

Dyson failed to prove that Intertek’s tests were not independent, that Dyson did not attack the tests’ compliance with ASTM F608 in any other way, and that Dyson’s expert “admitted that Intertek followed the proper procedure set forth in the controlling standard, ASTM F608.” Doc. 717 at 4. After Dyson’s response asserted that Intertek’s tests failed to comply with ASTM F608 because Intertek “used the wrong suction setting” and improperly selected vacuum samples, Doc. 723 at 13-14, SharkNinja addressed those matters in its reply, Doc. 727 at 1-3. Under these circumstances—where SharkNinja raised its understanding of the ASTM F608 compliance issue in its Rule 50(a) motion—SharkNinja preserved its challenge to the evidence regarding that issue. See Andy Mohr Truck Ctr., 869 F.3d at 605 (affirming the district court’s decision to consider a Rule 50(b) motion that “advanc[ed] a new argument in support of a ground

that appeared in” the Rule 50(a) motion, explaining that “[t]he district court reasonably found that [the movant] was not trying to slip a new point into the case”). By contrast, SharkNinja did not preserve its challenge to the sufficiency of the evidence regarding the grams graphic. True, SharkNinja’s opposition to Dyson’s Rule 50(a) motion challenged the evidence regarding the grams graphic. Doc. 724 at 3-10. But challenging the sufficiency of the evidence in response to a different motion—even the opposing party’s Rule 50(a) motion—is insufficient because Rule 50 requires the movant to first make (Rule 50(a)) and then renew (Rule 50(b)) such a challenge. See Fed. R. Civ. P. 50(b) (“If the court does not grant a motion for judgment as a matter of law made under Rule 50(a), the court is considered to have submitted the action to the jury subject to the court’s later deciding the legal questions raised by the motion. No later than 28 days after the entry of judgment[,] … the movant may file a renewed motion for judgment as a matter of law … .”) (emphasis added); see also Brown v. Smith, 827 F.3d 609, 613-14 (7th Cir. 2016) (holding that a summary judgment

motion does not preserve sufficiency of the evidence challenges for appeal, and that Rule 50(a) and 50(b) motions are required instead); Hammer v. Residential Credit Sols., Inc., 2015 WL 7776807, at *4 (N.D. Ill. Dec. 3, 2015) (“[T]he fact that [the movant] may have argued the timing issue at various other points in the proceedings is not sufficient to preserve the issue for [its] Rule 50(b) motion.”). Challenges made elsewhere to the sufficiency of the evidence, by definition, are not “the legal questions raised by the [Rule 50(a)] motion,” Fed. R. Civ. P. 50(b), and thus may not be brought in a Rule 50(b) motion. SharkNinja’s reply in support of its Rule 50(a) motion addressed the grams graphic issue by incorporating the arguments it had made in its response to Dyson’s Rule 50(a) motion. Doc. 727 at 1-3. But a submission that the movant makes for the first time in a reply brief is

forfeited for the purposes of that motion. See Narducci v. Moore, 572 F.3d 313, 324 (7th Cir. 2009) (“[T]he district court is entitled to find that an argument raised for the first time in a reply brief is forfeited.”); Cromeens, Holloman, Sibert, Inc. v. AB Volvo, 349 F.3d 376, 389 (7th Cir. 2003) (“Because Volvo raised the applicability of the Maine statute in its reply brief, the district court was entitled to find that Volvo waived the issue.”). And a ground forfeited for purposes of a Rule 50(a) motion cannot possibly qualify as a “ground[] … advanced in the preverdict [Rule 50(a)] motion” for purposes of determining which grounds have been preserved for a Rule 50(b) motion. Andy Mohr Truck Ctr., 869 F.3d at 604. Because it was not properly presented in SharkNinja’s Rule 50(a) motion, SharkNinja cannot raise in its Rule 50(b) motion any challenge to the sufficiency of the evidence regarding the grams graphic’s falsity. See id. at 604-05; Avery, 847 F.3d at 438. And because the grams graphic’s falsity, standing alone, supports the jury’s general verdict independently of Dyson’s

other theories, SharkNinja is not entitled to judgment as a matter of law as to liability. See Kossman v. Ne. Ill. Reg’l Commuter R.R. Corp., 211 F.3d 1031, 1037 (7th Cir. 2000) (“[W]hen a jury only returns a general verdict, we need only find support in the record for one of the theories presented to the jury in order to affirm the jury award.”); Culli v. Marathon Petroleum Co., 862 F.2d 119, 123 (7th Cir. 1988) (same). In any event, as shown below, SharkNinja’s grams graphic argument fails on the merits. B. Liability SharkNinja argues that Dyson failed to present evidence sufficient to support the jury’s verdict finding it liable for false advertising in violation of the Lanham Act. Doc. 754 at 7-15. The Lanham Act’s false advertising provision states in pertinent part: Any person who, on or in connection with any goods … , uses in commerce any word, term, name, symbol, or device, or any … false or misleading description of fact, or false or misleading representation of fact, which— … (B) in commercial advertising or promotion, misrepresents the nature, characteristics, [or] qualities … of his or her or another person’s goods … , shall be liable in a civil action by any person who believes that he or she is or is likely to be damaged by such act. 15 U.S.C. § 1125(a)(1). To prevail on a false advertising claim, the plaintiff must prove that “(1) the defendant made a material false statement of fact in a commercial advertisement; (2) the false statement actually deceived or had the tendency to deceive a substantial segment of its audience; and (3) the plaintiff has been or is likely to be injured as a result of the false statement.” Eli Lilly & Co. v. Arla Foods, Inc., 893 F.3d 375, 381-82 (7th Cir. 2018); see also Hot Wax, Inc. v. Turtle Wax, Inc., 191 F.3d 813, 819 (7th Cir. 1999) (similar). For purposes of the first element, precedent distinguishes between (1) statements that are literally false and (2) statements that are literally true or ambiguous but misleading. See Eli Lilly,

893 F.3d at 382; Hot Wax, 191 F.3d at 820. If a statement is literally false, it “will necessarily deceive consumers, so extrinsic evidence of actual consumer confusion is not required” to satisfy the second element. Eli Lilly, 893 F.3d at 382. If a statement is literally true or ambiguous but misleading, “the plaintiff ordinarily must produce evidence of actual consumer confusion in order to carry its burden” on the second element. Ibid. Only literal falsity is at issue here because Dyson has abandoned any claim that SharkNinja’s advertising was merely misleading. 259 F. Supp. 3d at 832. The Seventh Circuit has characterized literally false statements as “bald-faced, egregious, undeniable, over the top.” Schering-Plough Healthcare Prods., Inc. v. Schwarz Pharma, Inc., 586 F.3d 500, 513 (7th Cir. 2009). The central question, though, is “whether the defendant made

an explicit representation of fact that on its face conflicts with reality.” Eli Lilly, 893 F.3d at 382. Put another way, “[t]he proper domain of ‘literal falsity’ as a doctrine that dispenses with proof that anyone was misled or likely to be misled is the patently false statement that means what it says to any linguistically competent person.” Schering-Plough, 586 F.3d at 513. “[T]he meaning of the alleged literal falsehood must be considered in context and with reference to the audience to which the statement is addressed.” Ibid.; see also Abbott Labs. v. Mead Johnson & Co., 971 F.2d 6, 13-14 (7th Cir. 1992) (holding that the defendant’s description of its product as “rice-based,” which in context was a “term of art used to describe oral electrolyte solutions made from rice grain powder,” was literally false because the product contained “rice syrup solids” but not “powdered whole rice”). Whether an advertisement conveys a given message and whether that message is false are questions of fact. See Abbott Labs., 971 F.2d at 13-14 (holding that what message a product name conveys, whether the message is express or implied, and whether the message is false are questions of fact).

1. The Grams Graphic SharkNinja argues that Dyson failed to prove that the infomercial’s grams graphic scene was literally false, reasoning that because the infomercial combined a true voiceover (asserting that independent lab tests proved that SharkNinja’s vacuum cleans carpets better than Dyson’s) with a true bar graph (accurately displaying the results of SharkNinja’s internal testing), it was “(at worst) ambiguous.” Doc. 754 at 13-15 (capitalization altered). SharkNinja also argues that Dyson failed to prove “injury or materiality” as to the grams graphic. Id. at 15-16 (capitalization altered). Dyson responds that the grams graphic scene, viewed as a whole, unambiguously makes the false claim that the bar graph displays results from the independent tests discussed in the voiceover, and that the jury had ample evidence to conclude that the grams graphic was

material and harmful. Doc. 768 at 13-15. SharkNinja’s literal falsity argument rests on the incorrect premise that true plus true cannot equal false. When two statements are presented together, they contextualize each other and can thus communicate a different message in combination than they might in isolation. That is, the combination of two otherwise true statements can give rise to a third statement—a false one asserting that the first two are related. That is exactly what happens in the infomercial. As SharkNinja conceded at oral argument, the grams graphic appears in the infomercial “at exactly the same time” that the narrator says, “independent lab tests prove.” Doc. 810 at 17; Trial Exh. 12 at 6:07-6:11, 10:04-10:06. Thus, the jury reasonably could have found that the infomercial is literally false in conveying that the grams graphic represents the results of independent tests. True enough, the link between the voiceover and the graphic is subtler than it would have been had the narrator said, “[T]his graphic shows the results of independent testing.” Doc. 774

at 15. But such a direct narration is not the only way to eliminate ambiguity; context works just as well. Suppose a car is approaching a traffic light. Viewed from the direction the car is traveling, the light is red, but the passenger says, “The light is green.” Taken in isolation, the statement is true: The light is green when viewed from the cross street. In context, however, the statement is unambiguously false. Just as the direction in which the car is traveling determines which side of the light the driver will understand the passenger to be referencing without the passenger’s saying so, so too does the timing of the voiceover relative to the grams graphic communicate that the graphic displays the results of independent tests without the narrator’s saying so—or so a reasonable jury could find. See Abbott Labs., 971 F.2d at 13-14 (holding that what an advertisement communicates is a factual question). Given SharkNinja’s admission that

the grams graphic does not in fact show the results of independent tests, the jury reasonably could have found that SharkNinja “made an explicit representation of fact that on its face conflicts with reality,” and thus that the grams graphic scene is literally false. Eli Lilly, 893 F.3d at 382. At oral argument, SharkNinja contended that a barely visible footnote to the grams graphic attributing the data to tests performed on a “multi-level carpet sample” introduces ambiguity because it contradicts the voiceover’s reference to independent tests conducted “on four of the most commonly owned carpet types in America.” Doc. 810 at 18-19; see Trial Exh. 12 at 6:07-6:18, 9:56-10:15; Doc. 754-4 at 12 (transcript of the infomercial). SharkNinja forfeited this argument by failing to raise it in its Rule 50(a) or Rule 50(b) briefs. See Harden v. Marion Cnty. Sheriff’s Dep’t, 799 F.3d 857, 863 (7th Cir. 2015) (“Arguments that are raised for the first time in oral argument are forfeited.”). The argument fails on the merits in any event. Even if a barely visible footnote contradicting a voiceover could inoculate an advertisement

against a finding of literal falsity—a dubious proposition at best—there is no contradiction here. SharkNinja suggests that consumers who “just heard [from the voiceover] that four carpet types were tested in the independent tests” would conclude from a footnote mentioning only “one of those carpet types” that the footnote and the voiceover refer to different tests. Doc. 810 at 17. But consumers could just as reasonably conclude that SharkNinja tested all four types but chose to display the results for only one—as evidenced by the fact that SharkNinja did exactly that in the grams graphic. Doc. 754-18 at 18 (presenting internal test results for the NV650 on four types of carpet, including the multilevel carpet test results used in the grams graphic); Doc. 754- 21 at 18 (same, for the DC65); see Doc. 797 at 74-79 (Karyn Medler, SharkNinja’s vice president of global testing, identifying those results as among those used to support the grams

graphic). The footnote therefore does not make the grams graphic scene ambiguous. SharkNinja next argues that even if Dyson proved literal falsity, it failed to prove materiality because it presented no evidence that the false statement about the grams graphic’s source “actually influenced any consumer buying decisions.” Doc. 754 at 15. That is not the standard for evaluating materiality. Under the Lanham Act, a statement is material if “it is likely to influence the purchasing decision.” Hot Wax, 191 F.3d at 819. The evidence at trial enabled the jury to reasonably conclude that SharkNinja’s false statement about the source of the grams graphic was material. SharkNinja’s CEO testified that the graphic was meant to “show visually the amount of dirt that we each pick up,” which he thought would make an impact on consumers. Doc. 803 at 27. Moreover, a SharkNinja internal document shows that SharkNinja decided during the infomercial editing process to stick with the multilevel carpet results used in the graphic rather than using plush carpet results showing the NV650 with a slightly larger margin of victory because of the “perception of 42g [for the NV650] vs. 37g [for the DC65]” for the

multilevel carpet test as opposed to 48.5 grams vs. 42.9 grams for the plush carpet test. Doc. 768-15 at 5. As the document explained: “We like us being in the 40’s and them being in the 30’s.” Ibid. Given the evidence showing that SharkNinja thought the grams graphic would be impactful to viewers, the jury reasonably could have found the same. And from that finding, it reasonably follows that SharkNinja’s credibility-enhancing representation that the graphic reflected independent test results was likewise material. SharkNinja also contends that the grams graphic is immaterial because replacing the data displayed “with data from Intertek’s [independent] testing reflecting the total grams from all of the carpets … would have had the same visual impact on consumers,” as the grams graphic showed a five-gram difference for the multilevel carpet test while the Intertek tests would have

shown a four-gram difference on all four carpets combined. Doc. 754 at 15-16 (emphasis omitted). The jury was not required to reach that conclusion, particularly given the evidence that, “due to perception,” SharkNinja preferred a five-gram difference with its vacuum in the 40s and Dyson’s in the 30s over the plush carpet results showing a larger difference (5.6 grams) but with both vacuums in the 40s. Doc. 768-15 at 5. Moreover, the jury reasonably could have found that the visual impact of the data displayed in the graphic was greater than the impact that would have resulted from using Intertek’s tests: The graphic used in the infomercial showed Dyson’s vacuum picking up only 88 percent (37 grams divided by 42 grams) of what SharkNinja’s vacuum did, Trial Exh. 12 at 6:07-6:18, 10:04-10:15, while a graphic constructed from Intertek’s tests on all four carpets would have shown Dyson’s vacuum picking up 98 percent (166.8 grams divided by 170.9 grams) of what SharkNinja’s vacuum did, making the difference appear much less significant, Doc. 754 at 15 n.5. Finally, SharkNinja suggests that Dyson failed to prove that it was injured by the grams

graphic, but SharkNinja conflates the injury element with materiality and makes no independent argument about it. Doc. 754 at 15-16. SharkNinja’s challenge to the injury element thus fails for the same reasons as its materiality argument. In sum, a reasonable jury could have concluded that the infomercial’s grams graphic scene amounted to false advertising in violation of the Lanham Act. Because this theory of falsity is sufficient to support the jury’s general verdict as to liability, it does not matter whether the verdict could also rest on the independence representation. See Kossman, 211 F.3d at 1037; Culli, 862 F.2d at 123. In any event, as shown below, the evidence was sufficient to support a finding that the independence representation was false advertising as well. 2. The Independence Representation

SharkNinja contends that no reasonable jury could have found that the independence representation was literally false, reasoning that its instructions to Intertek to use the NV650’s high-suction setting when testing all four types of carpets did not undercut Intertek’s independence because SharkNinja gave those instructions in response to Intertek’s properly seeking clarification when the NV650 instruction manual and user interface left it unsure which setting to use. Doc. 754 at 8-9. Dyson responds by pointing to evidence suggesting that Intertek was not in fact confused and that SharkNinja nonetheless “overrode Intertek’s independent judgment and demanded that Intertek test all carpets with high suction.” Doc. 768 at 8-9. Although it presents a closer question than the grams graphic, a reasonable jury could have found that SharkNinja effectively controlled the tests by dictating to Intertek which setting to use, thus rendering the tests not independent and the independence representation false. ASTM F608 directs testers to determine which vacuum setting to use for a test by “using the manufacturer’s specifications as provided in the instruction manual for each type of carpet.”

Doc. 768-13 at § 9.2.3.1. ASTM F608 further instructs testers to “[c]ontact the manufacturer if no instructions are given, or if the instructions are unclear or inadequate.” Ibid. Steve Reese, an Intertek test operator, testified that Intertek’s “initial interpretation” of the NV650 manual was that multilevel and level loop carpets should be tested in one setting (“Carpet”), while plush and shag carpets should be tested in another (“Turbo”). Doc. 799 at 16, 20-21. Reese also testified that Intertek began testing based on that understanding of the manual. Id. at 18. Despite that understanding, Mark Titus, another Intertek test operator, emailed Karyn Medler of SharkNinja for clarification about which setting to use. Doc. 768-7 at 2; Doc. 799 at 16. Noting that the NV650 had “carpet,” “turbo,” and “hard floor” settings, while the NV651 (a vacuum operationally equivalent to the NV650 and part of the “NV650 series” of vacuums,

Doc. 794 at 113) had “thick carpet,” “carpet,” and “hard floor,” Titus asked, “Do you want us to use the ‘carpet’ setting for all carpets, or change it to ‘turbo’ for plush and shag per the user manual?” Doc. 768-7 at 2. Medler replied, “All carpet cleaning should be in turbo or thick carpet setting.” Ibid. Reese then wrote that “[u]nfortunately we have tested the [level loop] and [multilevel carpets] using the ‘Carpet’ mode per the owner’s manual,” and that Intertek would therefore “go back and retest on the ‘Turbo’ mode for you for those two carpets.” Ibid. Intertek’s reports for the tests state: “Testing was performed using ‘Turbo’ mode setting per client’s request on all ASTM approved carpets.” Doc. 768-11 at 5 (emphasis added); Doc. 768-12 at 5 (same). Alexander Porter, Intertek’s global director of engineering, testified that such a note could be used to indicate a “deviation” from the ASTM F608 standard or a “clarification” needed to reproduce the test in the same way it was conducted. Doc. 807 at 26- 27. Porter acknowledged that when he was deposed as Intertek’s corporate representative, he testified that the note reflected a deviation, but he offered a different view at trial, claiming that

he was “just speaking quickly” at the deposition and that there was not in fact a deviation. Id. at 27, 58-59, 64. Presented with that evidence, the jury reasonably could have found that Intertek was not confused by the manual and that ultimately SharkNinja made key decisions about the suction settings used to conduct the tests, undermining their independence. True, SharkNinja points to evidence that may well have permitted the jury to come out the other way. A few days after sending his email indicating that Intertek had already tested two types of carpet on “Carpet” instead of “Turbo,” Reese emailed Medler with “a problem”: Two NV651 units had different handles, one matching the NV650’s “Carpet” and “Turbo” settings, and the other with “Thick Carpet” and “Carpet” settings instead, prompting Reese to ask, “Huh????” Doc. 754-28 at 2.

Medler responded that the problem was caused by “a last minute change to the handle labeling” and that Intertek should use “the center setting.” Ibid. Moreover, Intertek’s copy of the SharkNinja manual was potentially confusing in that it: (1) incorrectly flipped the instructions, directing consumers to use “Turbo” for high-pile carpets and “Carpet” for low-pile carpets instead of the other way around, Doc. 754-25 at 10; Doc. 754-26 at 2; Doc. 799 at 19-20; and (2) elsewhere directed consumers to use the nonexistent “Turbo Carpet” setting for high-pile carpets, Doc. 754-25 at 10. But SharkNinja cannot win its Rule 50(b) motion by showing that the jury reasonably could have reached a different verdict; rather, it must show that the verdict the jury reached was unreasonable. See Fed. R. Civ. P. 50(a)(1), (b); Martinez, 900 F.3d at 844; Thorne, 882 F.3d at 644. Given the evidence discussed above, SharkNinja has not made that showing. Because the evidence concerning the grams graphic and the evidence concerning the independence representation independently suffice to support the liability verdict, it is

unnecessary to consider whether the evidence was sufficient to support a finding that SharkNinja’s advertisements were literally false because Intertek did not comply with ASTM F608 due to its use of the high-suction setting to test shag carpet, failure to randomly select the vacuums it tested, and failure to test a mass-produced NV650. See Kossman, 211 F.3d at 1037; Culli, 862 F.2d at 123. SharkNinja accordingly is not entitled to judgment as a matter of law as to its liability on Dyson’s false advertising claim. C. Damages SharkNinja argues in the alternative that even if the liability verdict stands, it is entitled to judgment as a matter of law because the evidence does not support any damage award. Doc. 754 at 16-20. Dyson sought an accounting of SharkNinja’s profits on the NV650 for the period from August 25, 2014 through December 15, 2014. See 15 U.S.C. § 1117(a) (authorizing an award of

“defendant’s profits”). The parties stipulated that SharkNinja’s gross profits from that period were $18,138,000. Doc. 809 at 51. From there, SharkNinja had the burden to prove “all elements of cost or deduction [it] claimed,” 15 U.S.C. § 1117(a), including any basis for apportioning profits to exclude those “not obtained through” its Lanham Act violation. WMS Gaming Inc. v. WPC Prods. Ltd., 542 F.3d 601, 608 (7th Cir. 2008); see also Roulo v. Russ Berrie & Co., 886 F.2d 931, 941 (7th Cir. 1989) (same). The jury awarded Dyson $16,410,681. Doc. 737. The jury appears to have calculated that award by starting with the $18,138,000 in gross profits and deducting the $1,727,319 in non- advertising expenses (commissions, deliveries, customer service call centers, warehouses and fulfillment, credit card processing fees, and bad debt) identified by SharkNinja’s economics expert. Doc. 754-31 at 2. In challenging the damage award, SharkNinja contends that (1) unrefuted evidence showed that, due to its substantial advertising expenses, it lost money on the NV650 during the relevant period and thus had no profits to disgorge, and (2) even if it made

a profit, the jury irrationally failed to apportion the profits to award Dyson only the share attributable to false advertising. Doc. 754 at 16-20. As to SharkNinja’s argument that it earned no profit during the relevant period, the jury was not required to believe SharkNinja’s evidence regarding its claimed advertising expenses, and thus reasonably could have found that SharkNinja did not meet its burden to prove that its gross profits figure should be reduced by those expenses. SharkNinja CFO Brian Lagarto admitted that the various SharkNinja financial documents shown to the jury, Docs. 768-17, 768-18, 768-19, 768-20, were created for this litigation because the company does not ordinarily prepare product-specific financial statements, Doc. 807 at 100-102, 134. Significantly, those documents provided vastly different accounts of SharkNinja’s NV650-related advertising

expenses and profits or losses during the relevant period. Doc. 807 at 115, 121-123 (Lagarto acknowledging those differences, including one document reporting a $30.6 million net profit versus two others reporting net losses of $14 million and $16 million). And Lagarto admitted that the documents contained substantial errors. Id. at 118 (Lagarto testifying that one document, Doc. 768-18, was inaccurate because it duplicated about $45 million in “infomercial revenue” and was missing about $33 million in “short-form media” expenses). Additionally, Dyson’s accounting and finance expert, Julie Davis—whom the jury was entitled to believe—disputed the accuracy and reliability of SharkNinja’s financial documents. Doc. 800 at 63-72. For example, Davis noted that one SharkNinja profit-and-loss statement included data for two more months than a second document yet reflected fewer total sales. Id. at 68-69. Davis observed that SharkNinja’s “variable selling expenses … changed rather dramatically” across the documents and opined that it was “concerning that we now have four sets of financial data, none of which agree with each other,” making it “hard to know which, if

any, of them are accurate.” Id. at 71-72. Given these concerns, Davis further opined that the documents did not provide a “sufficient [basis] to subtract any amount of media expense” for the NV650. Id. at 78; see also id. at 80 (“We know they spent money. We just don’t really have a reason to believe it was this amount of money.”). The jury thus was free to disbelieve SharkNinja’s fuzzy and inconsistent accounting and find that it did not prove any of its claimed advertising expenses as deductions from its gross profits, much less that its overall profit was zero. See Reeves, 530 U.S. at 151 (holding that the court must credit evidence favoring the Rule 50(b) movant only if the evidence is “uncontradicted and unimpeached” and “comes from disinterested witnesses”) (internal quotation marks omitted); Robinson, 894 F.3d at 833 (same). That leaves SharkNinja’s apportionment argument. From the premises that (1) not all its

profits on the NV650 resulted from its false advertising, and (2) apportionment is required under those circumstances, SharkNinja concludes that, as a matter of law, the verdict cannot have exceeded $0. Doc. 754 at 19-20. SharkNinja fails to explain how it gets from (a) “apportionment was required” because it came by some of its profits legitimately, id. at 20, to (b) a $0 award, which would mean that none of its profits resulted from its false advertising. (To the extent SharkNinja means to argue that the combination of deducting advertising expenses and apportioning profits gets it to zero, ibid., it runs into the same problem. An apportionment calculation produces $0 in damages only if the profits to be apportioned are already zero or if 100 percent of the profits came from other sources.) And because SharkNinja offers no middle ground between $16,410,681 and $0, it forfeits any argument that it is entitled as a matter of law to a middle-ground outcome. See G & S Holdings LLC v. Cont’l Cas. Co., 697 F.3d 534, 538 (7th Cir. 2012) (“We have repeatedly held that a party waives an argument by failing to make it before the district court.”); Alioto v. Town of Lisbon, 651 F.3d 715, 721 (7th Cir. 2011) (“We

apply [the forfeiture] rule where a party fails to develop arguments related to a discrete issue … .”). SharkNinja pursued much the same strategy at trial, declining to offer a view as to precisely how the jury should apportion profits should it find that SharkNinja earned profits during the relevant period—that is, SharkNinja did not make a backup argument along the lines of, “If you disagree with us that our profits were zero, then you should find that only X percent of our profits resulted from the false advertising, and enter an award in that amount.” Instead, SharkNinja urged the jury to award $0, period, full stop. That was an understandable strategy; SharkNinja may have feared that the jury, if presented with a backup argument, would split the baby rather than send Dyson home with nothing. SharkNinja’s strategy succeeded in part—the

jury did not split the baby—and blew up in part—the jury gave Dyson the entire baby. Having gambled and lost, SharkNinja “must live with the consequences of its strategic decision.” Int’l Coll. of Surgeons v. City of Chicago, 153 F.3d 356, 366 (7th Cir. 1998); see also Crowe ex rel. Crowe v. Zeigler Coal Co., 646 F.3d 435, 444 (7th Cir. 2011) (“Travelers … made a tactical decision not to intervene in this matter until the eleventh hour (at least), and now Travelers will have to live with the consequences of that decision.”); Illinois v. City of Chicago, 2018 WL 3920816, at *11 (N.D. Ill. Aug. 16, 2018) (holding that a party “must live with the consequences” of its “strategic” decisions), aff’d, 912 F.3d 979 (7th Cir. 2019). II. SharkNinja’s Motion for New Trial or Remittitur In the alternative to judgment as a matter of law under Rule 50(b), SharkNinja moves under Rule 59 for a new trial or remittitur. Doc. 756. In resolving a Rule 59(a) new trial motion, “federal law requires a district court to determine whether the verdict is against the weight of the evidence[,] the damages are excessive, or for other reasons, the trial was not fair to the party

moving.” Kapelanski v. Johnson, 390 F.3d 525, 530 (7th Cir. 2004) (alterations and internal quotation marks omitted); see also Pickett v. Sheridan Health Care Ctr., 610 F.3d 434, 440 (7th Cir. 2010) (same). SharkNinja’s motion for remittitur arises under Rule 59(e). See Stragapede v. City of Evanston, 865 F.3d 861, 865, 868 (7th Cir. 2017). “Altering or amending a judgment under Rule 59(e) is permissible when … there has been a manifest error of law or fact.” Harrington v. City of Chicago, 433 F.3d 542, 546 (7th Cir. 2006). A. Liability Verdict SharkNinja seeks a new trial on the ground that the liability verdict was against the weight of the evidence. Doc. 756 at 7-8. The court “is permitted to grant a new trial because the jury’s verdict was against the weight of the evidence only when the record shows that the jury’s

verdict resulted in a miscarriage of justice or where the verdict cries out to be overturned or shocks [the] conscience.” Prime Choice Servs., Inc. v. Schneider Logistics Transloading & Distribution, Inc., 861 F.3d 633, 635 (7th Cir. 2017) (alteration and internal quotation marks omitted); see also Estate of Burford v. Accounting Practice Sales, Inc., 851 F.3d 641, 646 (7th Cir. 2017) (same). “When considering whether the jury’s verdict goes against the manifest weight of the evidence, a court analyzes the general sense of the evidence, assessing the credibility of the witnesses and the comparative strength of the facts put forth at trial.” Willis v. Lepine, 687 F.3d 826, 836 (7th Cir. 2012) (internal quotation marks omitted); see also Mejia v. Cook Cnty., 650 F.3d 631, 633 (7th Cir. 2011) (same). “Jury verdicts deserve particular deference in cases with simple issues but highly disputed facts.” Moore ex rel. Estate of Grady v. Tuleja, 546 F.3d 423, 427 (7th Cir. 2008) (internal quotation marks omitted); see also Galvan v. Norberg, 678 F.3d 581, 588 (7th Cir. 2012) (“The district court also has less freedom to overturn a jury verdict in cases involving issues that are easily understood by laypeople.”).

SharkNinja’s argument under Rule 59(a) that the verdict is against the weight of the evidence merely incorporates its Rule 50(b) arguments. Doc. 756 at 7. Although “a motion for a new trial may be granted even if a motion for judgment as a matter of law must be denied,” Mejia, 650 F.3d at 634, that outcome is inappropriate here because, as shown above when addressing the Rule 50(b) motion, the liability verdict is well-supported by the evidence. See Williams v. Liefer, 491 F.3d 710, 716 (7th Cir. 2007) (holding that where “the jury had a reasonable basis in the record for returning a verdict in favor of [the plaintiff]” and the district court correctly denied the defendants’ motion for judgment as a matter of law, the district court did not abuse its discretion by also denying the defendants’ new trial motion challenging the verdict as against the weight of the evidence).

SharkNinja contends that so long as at least one of Dyson’s grounds for finding the advertisements literally false is unsupported, it is entitled to a new trial because the general verdict prevents the court from determining whether the jury relied on a “legally deficient” ground. Doc. 756 at 7-8; see Doc. 737 at 1 (verdict form asking, “Did Dyson establish by a preponderance of the evidence that SharkNinja engaged in false advertising for its Shark Rotator Powered Lift-Away from August 25, 2014 through December 15, 2014?”). Precisely the opposite is true: “[W]hen a jury only returns a general verdict, [the court] need only find support in the record for one of the theories presented to the jury in order to affirm the jury award.” Kossman, 211 F.3d at 1037; see also Culli, 862 F.2d at 123 (“[B]ecause the plaintiffs could have succeeded under one of two alternative independent theories of relief, and the jury returned a general verdict as opposed to a special verdict which, for example, could have included specific findings with respect to each theory of relief, we need only find that there was substantial evidence to support the jury’s verdict under at least one of the two alternative theories.”). That is

particularly so where, as here, the movant “never requested any special form of verdict.” Kossman, 211 F.3d at 1037. As the court observed at trial, neither SharkNinja nor Dyson “suggested any special interrogatories as to what the jury believes about the three or so theories of why the advertising was false.” Doc. 806 at 131. Indeed, when the court offered the option of asking the jury to answer special interrogatories, SharkNinja declined. Id. at 131-133. SharkNinja next reframes its true-plus-true-cannot-equal-false challenge to the grams graphic theory—rejected above in the context of its Rule 50(b) motion—as an argument that the “verdict is premised on … a legal impossibility.” Doc. 773 at 7-8. That argument does not persuade for two reasons. First, as noted in denying the Rule 50(b) motion, SharkNinja’s challenge to the grams graphic theory fails on the merits, as the jury had ample grounds to find

the infomercial’s grams graphic scene literally false. See Abbott Labs., 971 F.2d at 13-14. Second, the cases cited by SharkNinja to support its argument that one faulty theory invalidates a general verdict govern only those circumstances where instructional errors “left it open for the jury to base [its] verdict” on a legally incorrect theory. Sunkist Growers, Inc. v. Winckler & Smith Citrus Prods. Co., 370 U.S. 19, 25-26, 29-30 (1962) (holding that a new trial was required where the instructions incorrectly permitted the jury to find a conspiracy among parties who were legally incapable of forming a conspiracy); see also E. Trading Co. v. Refco, Inc., 229 F.3d 617, 622 (7th Cir. 2000) (interpreting Sunkist Growers as holding that a new trial is required where “the jury [was] instructed on an erroneous theory of liability,” “there [was] no basis for determining whether it relied on that theory,” and the erroneous theory was “supported by the facts”); Lawndale Nat’l Bank v. Am. Cas. Co. of Reading, 489 F.2d 1384, 1386, 1388-89 (7th Cir. 1973) (holding that a new trial was required where the jury was instructed on two defenses, one of which was precluded by a statute, and it was impossible to determine whether

the verdict was based on the precluded defense); Tex. Advanced Optoelectronic Sols., Inc. v. Renesas Elecs. Am., Inc., 895 F.3d 1304, 1316 (Fed. Cir. 2018) (distinguishing between liability theories that are “unsupported by sufficient evidence” and those that are “legally erroneous,” and noting that where a jury is instructed on a legally erroneous theory, a new trial is required “unless [the court] can tell that the jury came to its decision using only correct legal theories” or the error was harmless) (internal quotation marks omitted), petition for cert. filed, No. 18-600 (U.S. Nov. 7, 2018). Here, by contrast, Dyson’s grounds for liability are factual theories (various allegedly false statements made in SharkNinja’s advertisements) directed at a single legal theory (SharkNinja violated the Lanham Act by making literally false statements in advertising), and SharkNinja does not articulate a challenge to the jury instructions. Doc. 773 at 9 (SharkNinja

disputing Dyson’s contention that it conceded that the jury was properly instructed, but identifying no flaw in the instructions). The rule requiring a new trial in cases involving instructional errors therefore does not apply. See United States v. Townsend, 924 F.2d 1385, 1414 (7th Cir. 1991) (“It is one thing to negate a verdict that, while supported by evidence, may have been based on an erroneous view of the law; it is another to do so merely on the chance— remote, it seems to us—that the jury convicted on a ground that was not supported by adequate evidence when there existed alternative grounds for which the evidence was sufficient.”). SharkNinja next argues that a new trial is warranted because “the jury failed to follow the court’s instructions.” Doc. 756 at 8 (capitalization altered). Juries are presumed to understand and follow their instructions. See Evans v. Michigan, 568 U.S. 313, 328 (2013); CSX Transp., Inc. v. Hensley, 556 U.S. 838, 841 (2009); United States v. Mokol, 646 F.3d 479, 487 (7th Cir. 2011). SharkNinja’s sole basis for arguing that the jury failed to do so here is that SharkNinja disagrees with the jury’s verdict. Doc. 756 at 9 (“To the extent that the jury’s verdict of liability

might have rested on the jury’s mistaken belief that, e.g., the Grams Graphic was literally false … , a new trial should be ordered … .”). SharkNinja’s disagreement with a verdict amply supported by the evidence does not rebut the presumption that the jury followed its instructions, and thus does not warrant a new trial. SharkNinja also argues that the verdict cannot rest on the finding that the independence representation was literally false from August 25, 2014 through December 15, 2014 because that finding cannot be reconciled with the court’s holding on summary judgment that the independence representation was literally true after SharkNinja amended its instruction manual in December 2014. Id. at 9-10. SharkNinja’s own brief identifies the flaw in its logic: “[O]nce the errors in the manual were corrected and instructions clarified in December 2014,

SharkNinja’s position was beyond genuine dispute … .” Doc. 773 at 12. By arguing at summary judgment that the December 2014 manual amendment made its independence representation true “beyond genuine dispute,” SharkNinja necessarily conceded the possibility that there was a genuine dispute before that time. There is nothing inconsistent about the result reached by the court on summary judgment and the jury at trial: The independence representation was literally false from August 25, 2014 through December 15, 2014 because SharkNinja overrode Intertek’s independent judgment as to which suction setting the manual instructed consumers to use; in December 2014, SharkNinja amended the manual to prescribe its preferred setting; and that amendment made the independence representation true from that point on. 259 F. Supp. 3d at 833-35 (holding that the December 2014 manual amendment eliminated any concerns that SharkNinja’s instructions to Intertek about which suction setting to use rendered Intertek’s tests non-independent); see also Doc. 811 at 2-3. Finally, SharkNinja challenges the jury’s willfulness finding as against the weight of the

evidence, reasoning that its witnesses testified that the grams graphic was a placeholder accidentally left in the infomercial. Doc. 756 at 10. SharkNinja’s witnesses did offer that testimony, but the jury was free to disbelieve it. See United States v. Wilson, 879 F.3d 795, 802 (7th Cir. 2018) (“[I]t is the exclusive function of the jury to determine the credibility of witnesses, resolve evidentiary conflicts, and draw reasonable inferences.”) (internal quotation marks omitted); Venson v. Altamirano, 749 F.3d 641, 649 (7th Cir. 2014) (“[W]e cannot say that Venson’s credibility was so unscathed as to essentially compel the jury to credit his version of events over that of the defendants. … [T]he credibility of the parties’ competing accounts was for the jury to assess.”); see also Reeves, 530 U.S. at 151 (suggesting that the jury must believe evidence only if it is “uncontradicted and unimpeached” and “comes from disinterested

witnesses”) (internal quotation marks omitted). Instead, the jury reasonably could have found that when SharkNinja decided to keep the grams graphic in the infomercial because it “like[d] us [SharkNinja] being in the 40’s and them [Dyson] being in the 30’s,” it was at least reckless as to the truth of the voiceover’s representation that the graphic reflected independent testing. Doc. 768-15 at 5 (SharkNinja’s August 28, 2014 “show notes” for the infomercial documenting a recommendation to keep the multilevel carpet data instead of switching to plush carpet data “due to perception of 42g vs. 37g”); see Doc. 809 at 52 (instructing the jury that “SharkNinja acted willfully only if it knew that its advertising claim was false or if it acted with reckless disregard as to whether its advertising claim was false”). This is particularly so given SharkNinja’s knowledge that (1) the grams graphic used data from multilevel carpet, and (2) Intertek’s results showed SharkNinja’s NV650 losing to Dyson’s DC65 on multilevel carpet. Doc. 796 at 106-108 (Medler testifying that SharkNinja by August 2014 had Intertek’s reports showing that the Dyson DC65 beat the SharkNinja NV650 on multilevel carpet and that the infomercial ran after August

2014); see also Doc. 769 at 12-13 (Dyson setting forth other evidence casting doubt on SharkNinja’s version of events and suggesting that it acted willfully). Even if SharkNinja were correct that the evidence does not support a willfulness finding as to the grams graphic, the jury’s willfulness finding can rest solely on the independence representation. See Kossman, 211 F.3d at 1037; Culli, 862 F.2d at 123; Doc. 737 at 2 (verdict form asking, “Did Dyson establish by a preponderance of the evidence that SharkNinja engaged in intentional false advertising for its Shark Rotator Powered Lift-Away during the relevant time period?”). SharkNinja waited until its reply brief to address whether the evidence supported a finding that the independence representation was willfully false, thus forfeiting that issue. See Narducci, 572 F.3d at 324; Cromeens, 349 F.3d at 389.

B. Damages Verdict SharkNinja argues in the alternative that the $16,410,681 damage award is excessive because the jury (1) did not deduct its advertising expenses from its gross profits and (2) did not apportion its profits to award only that portion resulting from the false advertising. Doc. 756 at 11-17. SharkNinja suggests remitting the damage award to $1,413,282, which it calculates by multiplying its claimed “life-to-date incremental profit margin of 4.8%” by “the agreed net sales of $29,443,384 for the period of liability.” Id. at 14, 17. SharkNinja submits that this amount “does not even take into account the need to apportion the disgorgement award,” but it never suggests a figure that does so or even a ballpark apportionment rate. Ibid. Three factors guide the determination whether the jury’s award was excessive for purposes of a motion for a new trial or remittitur: (1) “whether the award is monstrously excessive,” (2) “whether there is a rational connection between the award and the evidence,” and (3) “whether the award is roughly comparable to awards made in similar cases.” Gracia v.

SigmaTron Int’l, Inc., 842 F.3d 1010, 1022 (7th Cir. 2016) (internal quotation marks omitted). SharkNinja does not address the third factor—likely because it is of little, if any, help in evaluating an accounting of profits—thus forfeiting any argument that the award is out of step with those made in similar cases. See G & S Holdings, 697 F.3d at 538; Alioto, 651 F.3d at 721. That leaves the first two factors, which “are really just two ways of describing the same inquiry: whether the jury verdict was irrational.” Adams v. City of Chicago, 798 F.3d 539, 543 (7th Cir. 2015). “In order to determine whether the jury’s verdict was irrational, the district court must review the trial record as a whole in the light most favorable to the verdict.” Ibid. SharkNinja’s argument that the jury irrationally failed to deduct advertising expenses from the gross profits figure fails for the reasons the court set forth above in rejecting its

Rule 50(b) argument. As noted, SharkNinja had the burden to prove deductions attributable to its expenses, and the jury was not required to believe its fuzzy and shifting financial documents. See 15 U.S.C. § 1117(a) (“In assessing profits the plaintiff shall be required to prove defendant’s sales only; defendant must prove all elements of cost or deduction claimed.”). The jury’s decision not to deduct SharkNinja’s claimed advertising expenses was therefore rational and does not support a remittitur or a new trial. See WMS Gaming, 542 F.3d at 609 (“Courts consistently find that when a trademark plaintiff offers evidence of infringing sales and the infringer fails to carry its statutory burden to offer evidence of deductions, the plaintiff’s entitlement to profits under the Lanham Act is equal to the infringer’s gross sales.”). As to apportionment, SharkNinja is correct that Dyson is not entitled to profits “not obtained through” SharkNinja’s Lanham Act violations. WMS Gaming, 542 F.3d at 608; see also Roulo, 886 F.2d at 941-42 (same). As with its expenses, SharkNinja had the burden to distinguish between profits attributable to its false advertising and profits attributable to

legitimate business activities. See 15 U.S.C. § 1117(a); WMS Gaming, 542 F.3d at 608. To satisfy that burden, SharkNinja had to offer evidence “sufficient to provide a fair basis of division so as to give the [plaintiff] all the profits that can be deemed to have resulted from the [violation].” Roulo, 886 F.2d at 941 (quoting Sheldon v. Metro-Goldwyn Pictures Corp., 309 U.S. 390, 402 (1940)). The apportionment of profits between lawful and unlawful activity “is highly fact-specific and should be left to the jury.” Bridgeport Music, Inc. v. Justin Combs Publ’g, 507 F.3d 470, 483 (6th Cir. 2007) (alterations and internal quotation marks omitted); see also Andreas v. Volkswagen of Am., Inc., 336 F.3d 789, 797-98 (8th Cir. 2003) (same). The trial record has evidence from which the jury could have concluded that SharkNinja’s profits for the NV650 did not result entirely from its false advertising. E.g.,

Doc. 808 at 21-22 (SharkNinja’s economics expert testifying that factors affecting demand for the NV650 could include price, customer reviews, word of mouth, brand reputation, and point- of-sale marketing); Doc. 800 at 90-92 (Dyson’s expert acknowledging the possibility that NV650 sales could have been driven in part by those factors and by the truthful portions of SharkNinja’s advertising). But SharkNinja made the strategic decision at trial to stand on its assertion that its profits were zero, rather than suggesting what share of any profits the jury found should be apportioned to the false advertising. Doc. 810 at 52 (SharkNinja conceding that “[w]e didn’t put a number on it”). Even now, SharkNinja insists that “apportionment was required” while refusing to suggest a percentage. Doc. 756 at 17; Doc. 810 at 48-50 (SharkNinja maintaining that Dyson “should not be getting 100 percent of our profits,” while declining to suggest a percentage between 100 and 0). It was not irrational for the jury, faced with SharkNinja’s unwillingness or inability to propose an apportionment, to decide that it could do no better. SharkNinja submits that the jury’s decision not to apportion its profits gave Dyson a

windfall. Doc. 756 at 17. Even if SharkNinja were right, that result accords with Seventh Circuit precedent recognizing that the proper apportionment “often cannot be ascertained with any reasonable certainty” and that the resulting windfall risk should be borne by the defendant, not the plaintiff. WMS Gaming, 542 F.3d at 607-08 (internal quotation marks omitted). As the Seventh Circuit explained: “There may well be a windfall to the trade-mark owner where it is impossible to isolate the profits which are attributable to the use of the infringing mark. But to hold otherwise would give the windfall to the wrongdoer.” Id. at 608 (quoting Mishawaka Rubber & Woolen Mfg. Co. v. S.S. Kresge Co., 316 U.S. 203, 206-07 (1942)); see also Nintendo of Am., Inc. v. Dragon Pac. Int’l, 40 F.3d 1007, 1012 (9th Cir. 1994) (holding that the district court did not abuse its discretion in refusing to apportion trademark damages where the

defendant’s suggested methods of apportionment were inadequate). Ultimately, and as noted above in the Rule 50(b) discussion, SharkNinja made a strategic choice to go all in on the apportionment issue, and it cannot now complain that it lost everything. See Int’l Coll. of Surgeons, 153 F.3d at 366; Crowe, 646 F.3d at 444; City of Chicago, 2018 WL 3920816, at *11. III. Dyson’s Motion for an Exceptional Case Finding Dyson moves for a finding that this case is exceptional under 15 U.S.C. § 1117(a), which states in pertinent part that “[t]he court in exceptional cases may award reasonable attorney fees to the prevailing party.” 15 U.S.C. § 1117(a). The Lanham Act “does not define ‘exceptional case.’” TE-TA-MA Truth Found.–Family of URI, Inc. v. World Church of the Creator, 392 F.3d 248, 257 (7th Cir. 2004). However, interpreting a materially identical provision in § 285 of the Patent Act, 35 U.S.C. § 285, the Supreme Court held that “an ‘exceptional’ case is simply one that stands out from others with respect to the substantive strength of a party’s litigating position (considering both the governing law and the facts of the case) or the unreasonable manner in

which the case was litigated.” Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545, 553-54 (2014). In so holding, the Court rejected the Federal Circuit’s test—requiring either “litigation-related misconduct of an independently sanctionable magnitude” or a finding that “the litigation was both brought in subjective bad faith and objectively baseless”—as “overly rigid” given the statute’s “inherently flexible” standard. Id. at 554-55 (internal quotation marks omitted). The parties agree that Octane Fitness’s understanding of “exceptional” applies to § 1117(a). Doc. 743 at 6-7; Doc. 761 at 8 & n.1; see Tobinick v. Novella, 884 F.3d 1110, 1117- 18 (11th Cir. 2018) (joining “[e]very circuit to have considered the issue” in holding that Octane Fitness applies to Lanham Act cases). “District courts may determine whether a case is ‘exceptional’ in the case-by-case

exercise of their discretion, considering the totality of the circumstances.” Octane Fitness, 572 U.S. at 554. In exercising its discretion, the court may look to “the comparable context of the Copyright Act,” for which Fogerty v. Fantasy, Inc., 510 U.S. 517 (1994), offered a “nonexclusive list of factors” that may be considered in deciding whether to award fees, “including frivolousness, motivation, objective unreasonableness (both in the factual and legal components of the case) and the need in particular circumstances to advance considerations of compensation and deterrence.” Octane Fitness, 572 U.S. at 554 & n.6 (internal quotation marks omitted). Ultimately, “[t]he decision whether to award Lanham Act attorney fees is left to the district court’s sound discretion.” Burford v. Accounting Practice Sales, Inc., 786 F.3d 582, 589 (7th Cir. 2015); see also Highmark Inc. v. Allcare Health Mgmt. Sys., Inc., 572 U.S. 559, 564 (2014) (“[T]he district court is better positioned to decide whether a case is exceptional because it lives with the case over a prolonged period of time.”) (citation and internal quotation marks omitted).

Dyson argues that this case is exceptional under § 1117(a) because: (1) the jury found that SharkNinja’s violation was willful; and (2) SharkNinja engaged in litigation misconduct. Doc. 743 at 5-6. SharkNinja counters that: (1) this was a mine-run case in which it mounted a reasonable defense; and (2) its litigation conduct, while not perfect, fell short of the sort of misconduct that makes a case exceptional. Doc. 761 at 7-8. A. Willful Trademark Infringement Dyson first argues that the jury’s willfulness finding is sufficient to support an exceptional case determination. Doc. 743 at 5-13. Although Dyson’s brief asserts that, “as a matter of law, where a defendant is found to be willfully in violation of the Lanham Act, that case is considered exceptional,” id. at 5 (citing BASF Corp. v. Old World Trading Co., 41 F.3d 1081, 1099 (7th Cir. 1994)), it correctly conceded at oral argument that an exceptional case

finding does not follow inexorably from a willfulness finding. Doc. 783 at 83-84; see Octane Fitness, 572 U.S. at 554 (“[T]here is no precise rule or formula for making these determinations … .”) (internal quotation marks omitted); Stryker Corp. v. Zimmer, Inc., 837 F.3d 1268, 1279 (Fed. Cir. 2016) (holding that under Octane Fitness’s totality of the circumstances test, “it does not necessarily follow” from a willfulness determination “that the case is exceptional”). SharkNinja submits that when the jury’s willfulness finding on the portion of Dyson’s claim that went to trial is considered along with all the other circumstances of this litigation, this case does not “stand[] out from others with respect to the substantive strength of [SharkNinja’s] litigating position.” Octane Fitness, 572 U.S. at 554. Given the particular circumstances of this case, SharkNinja has the better of the argument. Dyson’s sole basis for questioning the substantive strength of SharkNinja’s litigating position is the jury’s willfulness finding. Doc. 743 at 8-13 (rehashing the facts supporting the willfulness finding and asserting that it “renders this case exceptional”). Importantly, Dyson

does not contend that any of SharkNinja’s positions were “frivolous[],” “objective[ly] unreasonable[],” advanced in bad faith, or “exceptionally meritless,” Octane Fitness, 572 U.S. at 554 n.6, 555 (internal quotation marks omitted), or that SharkNinja “mount[ed] a scorched-earth defense to a meritorious claim in the hope of imposing prohibitive litigation costs on” Dyson, Nightingale Home Healthcare, Inc. v. Anodyne Therapy, LLC, 626 F.3d 958, 963 (7th Cir. 2010) (suggesting that the kind of defense making a case exceptional is one that “mirror[s]” the conduct that amounts to abuse of process: “predatory resistance to valid claims”). Indeed, Dyson concedes that SharkNinja did not “persist[] in a defenseless case in order to impose costs on Dyson.” Doc. 766 at 15. And although Dyson does assert that “the jury’s willfulness finding suggests that [SharkNinja’s] arguments were not reasonable,” id. at 14 (emphasis omitted), it

fails to develop or cite any authority for that argument, thus forfeiting the point. See M.G. Skinner & Assocs. Ins. Agency v. Norman-Spencer Agency, Inc., 845 F.3d 313, 321 (7th Cir. 2017) (“Perfunctory and undeveloped arguments are waived, as are arguments unsupported by legal authority.”). In the end, SharkNinja’s violation of the Lanham Act, though willful, was not “especially egregious.” Neuros Co. v. KTurbo, Inc., 698 F.3d 514, 521 (7th Cir. 2012). SharkNinja’s independence representation (which was false from August 25, 2014 through December 15, 2014) and grams graphic were just part of its advertising campaign for the NV650. The independence representation lent credibility to SharkNinja’s otherwise true assertion that lab tests showed that SharkNinja’s vacuum cleaned carpets better than Dyson’s. And the grams graphic was on the screen for only about twenty-four seconds of a twenty-two-minute infomercial. Trial Exh. 12 at 6:07-6:18, 10:04-10:15. Additionally, SharkNinja removed the grams graphic from the television version of the infomercial by October 27, 2014 and from the

online versions by December 2014. Doc. 743 at 12-13; Doc. 804 at 121, 133. Moreover, SharkNinja’s statements, while false, were not egregiously so. As noted, the grams graphic reflected internal test results but was misrepresented by the voiceover as reflecting independent test results. The results of both tests differed only in the details, as both supported (to varying degrees) the assertion that SharkNinja’s vacuum cleaned carpets better than Dyson’s. Trial Exh. 12 at 6:07-6:18 (displaying the grams graphic with a voiceover stating that “independent lab tests prove the new Rotator Powered Lift-Away has more suction power and actually deep cleans carpets better than the latest $600 Dyson”); Doc. 754-4 at 9 (transcript of the infomercial); 259 F. Supp. 3d at 832-35 (noting that Intertek’s tests on four types of carpet showed that SharkNinja’s vacuum “removed dirt better” than Dyson’s, and holding that

SharkNinja’s claim about what the test proved was therefore true); Doc. 796 at 107 (Medler testifying that the grams graphic was based on internal test results showing that SharkNinja beat Dyson on multilevel carpet); Doc. 797 at 74-81 (Medler explaining exhibits presenting those results, e.g., Doc. 754-18 at 18; Doc. 754-21 at 18); Docs. 754-7, 754-9 (Intertek test results showing that while SharkNinja beat Dyson when compiling the results for the four carpet types, SharkNinja lost the multilevel carpet test). The falsity of the independence representation from August through December 2014 was likewise in the details, as evidenced by the relatively straightforward fix to SharkNinja’s instruction manual that made the representation true: As explained in the court’s summary judgment ruling, all SharkNinja had to do was amend its manual to direct consumers (and thus the testers) to use the same high-suction setting it had told Intertek to use—a setting on which Intertek’s test showed that SharkNinja’s vacuum beat Dyson’s. 259 F. Supp. 3d at 832-33, 835; see also Doc. 811 at 2-3. This would have been a different case had SharkNinja run its advertisements without ever

testing its vacuum’s performance, or had SharkNinja said that the tests showed that its vacuum cleaned carpets better than Dyson’s when in fact they showed the opposite. See Neuros, 698 F.3d at 518, 521 (suggesting that an exceptional case finding is appropriate where the defendant’s false accusations that its competitor committed fraud were “based on computational errors and incorrect assumptions” and the defendant “persisted in [those] representations … even after … compelling evidence was presented that the representations were false”); BASF, 41 F.3d at 1083, 1099 (holding that the district court did not abuse its discretion in finding the case exceptional where the defendant “advertised that its antifreeze met certain industry specifications” even though it “had never actually tested its antifreeze to determine whether it met [those] specifications”). But SharkNinja’s violations of the Lanham Act were not of that

stripe. What is more, although Dyson prevailed on the portion of its Lanham Act claim that went to trial, the parties earned mixed results over the course of the entire litigation. Earlier in the case, the court granted SharkNinja partial summary judgment as to Dyson’s implied falsity claims, claims regarding the Ball Multi-Floor vacuum, and claims for the period beginning December 2014, 259 F. Supp. 3d at 836-37; in addition, the court denied Dyson’s preliminary injunction motion, 2015 WL 1120006, at *1. These mixed results weigh against a determination that this case “stands out from others with respect to the substantive strength of a party’s litigating position.” Octane Fitness, 572 U.S. at 554; see also B&B Hardware, Inc. v. Hargis Indus., Inc., 912 F.3d 445, 454 (8th Cir. 2018) (holding that the case was not exceptional where it had “arguable merit on both sides,” as “evidenced by the fact that both parties … prevailed at various times throughout its 12-year history”). In sum, given the facts and circumstances of the case, an exceptional case finding is

unwarranted. See Coty Inc. v. Excell Brands, LLC, 277 F. Supp. 3d 425, 469 (S.D.N.Y. 2017) (“[C]ourts routinely decline to award attorneys’ fees in cases involving willful infringement.”) (internal quotation marks omitted) (collecting cases). In so concluding, the court notes that its rejection of an exceptional case finding in part on the ground that the false portions of SharkNinja’s advertisements were only part of its overall campaign may appear inconsistent with its earlier holding that the jury reasonably declined to apportion profits. Any such tension is illusory. As an initial matter, Dyson has the burden of proof on the exceptional case issue, see Octane Fitness, 572 U.S. at 557, while SharkNinja had the burden on the apportionment issue and can upset the jury’s verdict only if it was irrational, Thorne, 882 F.3d at 644; WMS Gaming, 542 F.3d at 608. It follows that victory for SharkNinja on the exceptional case issue can be

reconciled with victory for Dyson on the apportionment issue. Moreover, as explained above, the jury’s apportionment decision was rational even if some of SharkNinja’s profits were attributable to other sources because SharkNinja made the strategic decision not to offer the jury any answer as to how the profits should be apportioned. B. SharkNinja’s Alleged Litigation Misconduct Dyson next argues that SharkNinja engaged in three forms of litigation misconduct that warrant an exceptional case determination: (1) SharkNinja violated its discovery obligations during the preliminary injunction proceedings; (2) SharkNinja’s witnesses gave false testimony; and (3) SharkNinja either “allowed the destruction of or attempted to destroy” Intertek test results it was obligated to preserve. Doc. 743 at 13-19 (capitalization altered). SharkNinja denies these charges and asserts that Dyson’s own alleged misconduct cuts against an exceptional case finding. Doc. 761 at 15-21. Dyson retorts that SharkNinja’s opposition brief amounts to further misconduct because it contains “half-truths” and wrongfully accuses Dyson of misconduct. Doc. 766 at 18-20.

As noted, a case that “stands out from others with respect to … the unreasonable manner in which the case was litigated” is exceptional. Octane Fitness, 572 U.S. at 554. Although litigation conduct need not be “independently sanctionable” to be deemed exceptional, it is the “rare case” in which non-sanctionable conduct will justify a fee award. Id. at 555. “Playing hard—by the rules—cannot suffice to make a case exceptional under § 1117(a).” TE-TA-MA, 392 F.3d at 264. Rather, as the Seventh Circuit has explained, litigation conduct crosses the line if it is “particularly egregious,” “oppressive,” or “vexatious.” Id. at 258, 261, 264; accord Nightingale, 626 F.3d at 962 (similar). The Seventh Circuit does not lightly invoke those characterizations of the losing party’s litigation conduct. See TE-TA-MA, 392 F.3d at 263-64 (holding that the defendant’s litigation conduct made the case exceptional where it “purposely

orchestrated a campaign of harassment throughout the litigation … to force the [plaintiff] and its attorneys to drop the trademark claim and to drive up the costs of litigating the case,” which was “particularly egregious because much of it consisted of explicit threats of violence,” including an “attempt[] to have the presiding district court judge murdered”); S Indus., Inc. v. Centra 2000, Inc., 249 F.3d 625, 627-29 (7th Cir. 2001) (affirming an exceptional case finding where the plaintiff’s “indefensible claims” alone would have supported a fee award, where the plaintiff “added to the cost and aggravation of this meritless litigation by not responding to discovery requests, repeatedly failing to properly serve or sign motions filed with the court, and failing to satisfy the requirements of the local rules of the district court,” and where the plaintiff’s actions “look[ed] to be part of a pattern of abusive and improper litigation” across dozens of cases); Gorenstein Enters., Inc. v. Quality Care-USA, Inc., 874 F.2d 431, 437 (7th Cir. 1989) (affirming an exceptional case finding where the defendants acted “in bad faith,” “committed perjury in answering interrogatories,” raised “ridiculous” defenses, and “attempt[ed] to circumvent the

district judge’s ruling” denying leave to amend their counterclaim “by filing a fresh lawsuit patently barred by res judicata”). SharkNinja’s litigation conduct was not perfect, but Dyson fails to show that SharkNinja’s shortcomings were exceptional. First, Dyson argues that SharkNinja deliberately withheld unfavorable documents from production during the preliminary injunction proceedings and refused to comply with a court order during that time frame requiring production of Intertek’s files. Doc. 743 at 13-16. The discovery at issue was conducted under a stipulated order providing that “[t]he parties will work in good faith to engage in limited expedited discovery necessary to prepare for the evidentiary hearing” on Dyson’s preliminary injunction motion. Doc. 72 at ¶ 2. The parties served document requests on December 26, 2014, with

responsive documents due January 3, 2015, in time for depositions beginning January 5 and an evidentiary hearing on January 13. Id. at ¶¶ 1, 6. The parties agreed to hold off on “formal, written responses” until after the hearing. Id. at ¶ 6. As Dyson noted in its post-preliminary injunction hearing discovery responses, the parties “specifically agreed that the … expedited discovery would not include documents identified by electronic searching.” Doc. 761-31 at ¶ 14. Moreover, both parties acknowledged that the expedited discovery process preceding the preliminary injunction hearing necessarily resulted in incomplete responses. Id. at ¶ 15 (Dyson noting that its “discovery and investigation” is “ongoing,” that its document production was “made in a preliminary and expedited manner” and therefore “limited to information obtained and reviewed to date,” and that it reserved the right to “amend or supplement” as its investigation continued “and additional documents and information are subsequently developed or discovered”); Doc. 761-32 at 2 (SharkNinja noting that its “objections and responses are based upon information and documents currently available,” “are made in light of the premature and

expedited nature of discovery in this case,” and “are given without prejudice to [its] right to produce or rely on subsequently discovered, uncovered or learned information and documents”). SharkNinja’s discovery responses during the preliminary injunction proceedings indeed turned out to be incomplete. Doc. 761 at 16-17 (SharkNinja implicitly acknowledging the point). Given the aggressive preliminary injunction discovery schedule, the parties’ joint decision that electronically stored information would not be searched, and their insistence that their responses remained incomplete, however, Dyson has not established that SharkNinja’s omissions were the result of a coverup and that it did not endeavor to “cooperate in good faith” as directed by the preliminary injunction discovery order. Doc. 72 at ¶ 8. Dyson’s contention that SharkNinja ignored the order requiring the production of

Intertek’s files, Doc. 743 at 15-16 (citing Doc. 72 at ¶ 6), is likewise unpersuasive. Dyson’s argument relies on a selective reading of a letter in which SharkNinja (through its former lawyers) acknowledged that it did not “collect[] or produce[] documents directly from Intertek files during expedited discovery.” Doc. 508-15 at 3. Dyson presents that statement as a “blithe[]” admission that SharkNinja simply refused to comply with the order. Doc. 743 at 15- 16. But the letter goes on to explain why SharkNinja did not collect or produce documents directly from Intertek’s files: SharkNinja determined that it already possessed copies of all responsive documents in Intertek’s possession, so it produced copies from its files but not duplicates from Intertek’s files. Doc. 508-15 at 3-4. SharkNinja pointed this out in its brief, Doc. 761 at 18, but Dyson’s reply fails to address the argument, instead doubling down on its mischaracterization of SharkNinja’s letter without identifying what it thinks SharkNinja left out of its production, Doc. 766 at 17. Dyson therefore forfeits any challenge to SharkNinja’s assertion that the documents from Intertek’s files were duplicates of documents that SharkNinja

produced. See Webb v. Frawley, 906 F.3d 569, 581 (7th Cir. 2018) (“Webb has waived any counterarguments he may have had by not responding to Frawley’s argument on this topic in his reply brief.”); Bonte v. U.S. Bank, N.A., 624 F.3d 461, 466 (7th Cir. 2010) (“Failure to respond to an argument … results in waiver.”). And even if the discovery order could be read to have required the production of duplicates, Dyson does not explain how SharkNinja’s failure to do so could have been prejudicial, Doc. 743 at 15-16; Doc. 766 at 17, thus undercutting the notion that its conduct was so egregious as to warrant a fee award. Second, Dyson asserts that SharkNinja witnesses “lied under oath.” Doc. 743 at 16. SharkNinja responds that “at most, a couple of [its] witnesses gave mistaken testimony based on incomplete information.” Doc. 761 at 18-19. Dyson calls that “an understatement,” asserting

that one trial witness (Medler) “unequivocally lied, regardless of her subjective intent.” Doc. 766 at 18 (emphasis added). Dyson’s argument defeats itself, as the difference between giving incorrect testimony and lying under oath is the witness’s subjective intent. See Lie, Black’s Law Dictionary (10th ed. 2014) (defining a “lie” as a “false statement or other indication that is made with knowledge of its falsity; an untruthful communication intended to deceive”). As the court instructed the jury when Dyson raised this issue at trial, it was the jury’s prerogative to decide whether Medler made “an honest mistake” or gave “intentionally incorrect” testimony. Doc. 801 at 96-97. But the court must resolve whether Dyson proved by a preponderance of the evidence that this was an exceptional case. See Highmark, 572 U.S. at 563-64 (holding that “the exceptional-case determination is … reviewed only for abuse of discretion” because “the text of the statute emphasizes the fact that the determination is for the district court” and because the inquiry is “rooted in factual determinations”) (internal quotation marks omitted); Octane Fitness, 572 U.S. at 557 (holding that a preponderance of the evidence standard applies to the Patent

Act’s identical provision). Between its implicit concession that Medler’s “subjective intent” may not have been to deceive and its failure to offer more than speculation to prove that her intent was improper, Dyson has not met that burden. Medler’s testimony therefore does not support an exceptional case finding. Dyson also argued at first that a second witness, Alexander Porter, lied in the deposition he gave before the preliminary injunction hearing, Doc. 743 at 16, but it abandoned that argument in its reply brief, Doc. 766 at 18, after SharkNinja pointed out that the January 12, 2015 email cited by Dyson does not establish that Porter lied when he testified on January 9, 2015 that he was not aware of the information provided in that email. Doc. 761 at 18; see Webb, 906 F.3d at 581; Bonte, 624 F.3d at 466. Dyson also forfeits its argument that Medler and Porter

“were not alone” in lying by failing to develop the point. See M.G. Skinner, 845 F.3d at 321. Third, Dyson argues that SharkNinja engaged in spoliation by “allow[ing]” Intertek to destroy relevant July 2014 test results and by attempting to destroy results from a January 2015 test it canceled but for which Intertek nonetheless sent a report. Doc. 743 at 18-19. SharkNinja responds that it did not have a legal right to obtain or preserve Intertek’s July 2014 data and thus was not responsible for its destruction, and that regardless of whether the January 2015 data was handled appropriately, Dyson suffered no prejudice because the data was not in fact destroyed and Dyson in fact was able to use it at trial. Doc. 761 at 19-20. Dyson’s reply brief fails to address either issue, except to assert in a footnote to an unrelated paragraph that “Intertek would have provided the [July 2014] data to [SharkNinja] had it been asked” and to vaguely charge that SharkNinja is not “absolve[d] … of the numerous instances of withholding and hiding incriminating evidence.” Doc. 766 at 18 n.12. Dyson fails to develop or cite legal authority for these positions, thus forfeiting them. See M.G. Skinner, 845 F.3d at 321. And by failing to

otherwise engage with SharkNinja’s response, Dyson forfeits its spoliation argument. See Webb, 906 F.3d at 581; Bonte, 624 F.3d at 466. Even setting aside forfeiture, Dyson’s arguments fail on the merits. As to the July 2014 test results, the court already rejected Dyson’s argument in the context of its request for a spoliation instruction: [T]he question is not whether the party in the case could have asked for the documents and the non-party would have provided the documents or preserved the documents in this case. The question is whether the party in the case had a legal right to the documents. And that showing has not been made. Dyson has not addressed, let alone demonstrated, that SharkNinja had the legal right to the documents possessed by Intertek. So … that necessary predicate of the spoliation issue … is not present. Doc. 809 at 16 (citing Chaveriat v. Williams Pipe Line Co., 11 F.3d 1420, 1426-27 (7th Cir. 1993)). Dyson has made no progress toward demonstrating that SharkNinja had the legal right to the Intertek materials in question. Dyson’s opening brief asserts that “Shark’s agreement with Intertek to conduct the tests allowed Shark to obtain the results of the tests Intertek ran on its behalf.” Doc. 743 at 18. But all it cites in support is a nonexistent page of the trial transcript. Compare ibid. (citing page 2885), with Doc. 809 at 172 (trial concludes on page 2739). SharkNinja guesses plausibly, Doc. 761 at 19, that Dyson meant to cite page 2285 of the transcript, but the testimony there establishes only that “if Shark had asked for [the] data, Intertek would have provided it,” Doc. 807 at 32. That leaves Dyson exactly where it was when the court rejected its spoliation argument at trial. Doc. 809 at 16. As to the January 2015 test results, Dyson cannot demonstrate prejudice, which leaves the court unconvinced that any misconduct regarding those results was exceptional. Finally, Dyson contends that SharkNinja’s brief is itself “proof that this case is exceptional” because, in Dyson’s telling, the brief “tell[s] half-truths” and wastes SharkNinja’s

own “time and resources” accusing Dyson of misconduct. Doc. 766 at 18-20. Yet Dyson’s sole example of a half-truth is its assertion that SharkNinja’s brief falsely implies that because the independence representation “became literally true in December 2014,” it must always have been literally true, while “leav[ing] out to [sic] the fact that it changed its manual” to make the representation true. Id. at 18-19 (emphasis omitted) (citing Doc. 761 at 7). The only half-truth is in Dyson’s account of what SharkNinja’s brief says. The cited passage of the brief has nothing to do with whether the independence representation was true before December 2014; rather, the brief argues that this case “does not stand out with respect to the substantive strength of Dyson’s litigating position” because SharkNinja won summary judgment as to the period beginning December 2014. Doc. 761 at 7.

Dyson inexplicably throws into its paragraph complaining about SharkNinja’s brief the argument that SharkNinja’s Rule 50(b) motion reverses its position as to how to evaluate the grams graphic. Doc. 766 at 19 (citing Doc. 754 at 6). Dyson cites no authority for the proposition that attorneys’ “shifting argument[s]” can make a case exceptional, ibid., thus forfeiting the point. See M.G. Skinner, 845 F.3d at 321. In any event, SharkNinja did not reverse its position. Its argument before trial that the jury should not see transcripts of the infomercial because “no consumer ever saw” them and they reflected “only part of the real-world ads,” Doc. 526 at 1, is not inconsistent with its (incorrect) argument now that the infomercial’s grams graphic scene cannot be literally false when viewed as a whole because it is the sum of true parts, Doc. 754 at 6. And as to Dyson’s complaint that SharkNinja accused it of misconduct, Dyson fails to explain how this behavior makes the case exceptional or to cite any authority supporting its position, thus forfeiting the point. See M.G. Skinner, 845 F.3d at 321. Ultimately, SharkNinja’s conduct in this case was the sort of “aggressive litigation

conduct undertaken in good faith and falling within the bounds of zealous advocacy,” not “the sort of egregious harassment … [that] clearly qualifies as oppressive.” TE-TA-MA, 392 F.3d at 264. Any missteps by SharkNinja qualify as mine-run, not exceptional. This is therefore not “the rare case in which a party’s unreasonable conduct … is … so ‘exceptional’ as to justify an award of fees.” Octane Fitness, 572 U.S. at 555. IV. Dyson’s Motion to Amend the Judgment to Include Prejudgment Interest Dyson seeks $2,251,202 in prejudgment interest on the $16,410,681 damage award. Doc. 750 at 1-2. SharkNinja responds that prejudgment interest is inappropriate because: (1) the award is based on disgorgement of its ill-gotten gains rather than on Dyson’s actual damages; (2) SharkNinja’s conduct was not outrageous; and (3) the jury’s damage award is excessive. Doc. 767 at 5-6. SharkNinja does not challenge Dyson’s calculation of the prejudgment interest

it would owe on a $16,410,681 award, id. at 2 n.2, thus forfeiting any such challenge. See G & S Holdings, 697 F.3d at 538; Alioto, 651 F.3d at 721. Whether to award prejudgment interest is a matter committed to the district court’s discretion. See Frey v. Coleman, 903 F.3d 671, 682 (7th Cir. 2018). That said, the Seventh Circuit has adopted a presumption in favor of awarding prejudgment interest “to victims of federal law violations” because, “[w]ithout it, compensation of the plaintiff is incomplete and the defendant has an incentive to delay.” Gorenstein, 874 F.2d at 436; see also Frey, 903 F.3d at 682 (“Courts award prejudgment interest because compensation deferred is compensation reduced by the time value of money, and only prejudgment interest can make the plaintiff whole.”) (internal quotation marks omitted); RK Co. v. See, 622 F.3d 846, 853 (7th Cir. 2010) (noting that the presumption “is well-established in this circuit”); McRoberts Software, Inc. v. Media 100, Inc., 329 F.3d 557, 572 (7th Cir. 2003) (“The rule we articulated in Gorenstein was broad and we have consistently applied the presumption in favor of prejudgment interest for

willful violations of federal law in the years since.”). “[P]rejudgment interest is not meant to penalize the party who caused the injury.” Thorncreek Apartments III, LLC v. Mick, 886 F.3d 626, 637 (7th Cir. 2018). Rather, it “is simply an ingredient of full compensation that corrects judgments for the time value of money.” Pickett, 813 F.3d at 646 (internal quotation marks omitted). SharkNinja contends that where, as here, “a verdict is based upon a disgorgement of profits, the rationale [for] awarding prejudgment interest to ensure full compensation for the plaintiff is inapplicable.” Doc. 767 at 4-6. The only Seventh Circuit authority SharkNinja cites in support is Marshall v. Security State Bank of Hamilton, 970 F.2d 383 (7th Cir. 1992), which held that prejudgment interest was inappropriate where the plaintiffs were awarded only statutory

damages, the plaintiffs suffered no actual damages, and the defendant was not “enriched,” reasoning that “the rationale for awarding … prejudgment interest—that otherwise compensation is incomplete and the defendant is unjustly enriched—does not apply in this case.” Id. at 384-85 (emphasis added). Because the purpose of awarding the defendant’s profits as damages is in part to return the defendant to the position it would have been in had it not unjustly enriched itself, Marshall in fact explains why Dyson’s request for prejudgment interest is consistent with the “rationale for awarding” that remedy. Id. at 385; see also Roulo, 886 F.2d at 941 (“Profits are awarded [in Lanham Act cases] under different rationales[,] including unjust enrichment, deterrence, and compensation.”). Taking away SharkNinja’s ill-gotten gains in 2014 dollars, but letting it keep the “economic return [it] made (or could have made) by investing” those funds between 2014 and the date of judgment, would leave it unjustly enriched. SEC v. Koenig, 557 F.3d 736, 745 (7th Cir. 2009) (“Koenig’s ‘pecuniary gain’ is the amount he obtained by his fraudulent accounting,

plus the economic return he made (or could have made) by investing that sum between [the time he obtained it] and the date of disgorgement. And prejudgment interest is the right way to estimate the second component. Depriving Koenig of both the principal amount, and the economic return measured by prejudgment interest, puts him in the same position as if he had not received any ill-got gains … .”). Just as with actual damages, awarding prejudgment interest on SharkNinja’s profits would “simply … correct[ the] judgment[] for the time value of money,” Pickett, 813 F.3d at 646, and thus eliminate “an incentive to delay,” Gorenstein, 874 F.2d at 436. The fact that this case involves an award of SharkNinja’s profits therefore does not rebut the presumption in favor of prejudgment interest. SharkNinja next argues that prejudgment interest is inappropriate because it won partial

summary judgment, “defended itself in good faith,” and did not engage in “outrageous, flagrant or malicious” conduct. Doc. 767 at 7-8 (capitalization altered). This argument rests on Gorenstein’s observation that prejudgment interest was “particularly appropriate” in that case because “the violation was intentional, and indeed outrageous.” 874 F.2d at 436. But as noted, Gorenstein does not limit prejudgment interest to “outrageous” cases; to the contrary, it holds that “prejudgment interest should be presumptively available to victims of federal law violations.” Ibid. (emphasis added). It would make no sense to allow SharkNinja to rebut that presumption merely by showing that it engaged only in a mine-run violation; presumptions are meant to provide answers in mine-run cases, and Gorenstein’s answer is that courts should award prejudgment interest. See ibid. Moreover, none of the considerations invoked by SharkNinja changes the blunt economic reality underlying prejudgment interest: “Given inflation and the power of money to earn an economic return, a dollar received in [2014] is worth considerably more than a dollar” now. Koenig, 557 F.3d at 745. Finally, SharkNinja contends that prejudgment interest would “give Dyson an unjustified windfall” because the jury’s verdict was “excessive and unsustainable.” Doc. 767 at 3-4. This argument merely retreads SharkNinja’s direct challenges to the damage award in its Rule 50(b) and Rule 59 motions, Docs. 754, 756, and it fails for the same reasons that those challenges failed. Moreover, to the extent SharkNinja means to argue that a verdict’s being excessive warrants denying prejudgment interest altogether rather than calculating prejudgment interest based on an appropriate amount of principal, the argument is forfeited because SharkNinja cites no supporting authority. See M.G. Skinner, 845 F.3d at 321. Conclusion SharkNinja’s motions for judgment as a matter of law and for a new trial or remittitur are denied, as is Dyson’s motion for a finding that this case is exceptional. Dyson’s motion for prejudgment interest is granted. The judgment is amended to add $2,251,202 in prejudgment interest, making the total award $18,661,883.

March 31, 2019 United States District Judge